Northeastern Naturalist 392 T. Bookout and G.L. Bruland 22001199 NORTHEASTERN NATURALIST 2V6(o2l). :2369,2 N–4o0. 92 Assessment of a Restored Wetland in West-Central Illinois Taylor Bookout1,2,3,* and Gregory L. Bruland1 Abstract - Although natural wetlands provide important ecosystem services such as flood control, carbon sequestration, and habitat for wetland plants and amphibians, it is uncertain to what degree restored wetlands provide these services. To this end, we assessed the hydrology, soils, vegetation, and anuran relative call frequency in a restored emergent floodplain wetland in west-central, Illinois. We employed a stratified random design to sample across a hydrologic gradient from wetter to drier zones in 3 cells of the wetland. We monitored surface water levels and found that cells 1 and 2 showed long periods of inundation, while cell 3 exhibited a more pulsed hydrology based on rainfall. Soil moisture content exhibited a significant trend across the hydrologic gradient, increasing from the drier to the wetter zones. We identified 46 plant species, 14 of which were planted as part of the restoration. Plant communities differed among cells, with cells 1 and 2 having more than 37% obligate wetland species, while cell 3 had only 22% obligate wetland species. Over 30 survey nights, we heard 10 anuran species calling and observed 1 Ambystoma (mole salamander). Hydrology played an important role in site usage by amphibians, especially in cell 3, where the absence of water precluded egg laying. Introduction Natural wetlands offer important ecosystem services including floodwater retention, carbon sequestration, and habitat for threatened and endangered species (Mitsch and Gosselink 2015). As the importance of wetlands and their services to humans have become more widely recognized, state and federal policies have shifted from wetland conversion to protection and restoration. The Clean Water Act (1972), was a turning point in US wetland policy and requires mitigation whenever wetlands are dredged or filled. This regulation has resulted in the restoration and creation of wetlands to offset losses to development. These wetland restoration projects are monitored by state and federal agencies to determine the success of the restoration. Vegetation and hydrology are the most commonly used metrics to assess the ecological status of restored wetlands (Matthews and Endress 2008). Recently, soils (Ballantine and Schneider 2009, Bantilan-Smith et al. 2009) and amphibian usage have also been studied as important indicators of wetland restoration success (Pillsbury and Miller 2008, Shirose et al. 1997). Wetland soils are the foundation of wetland ecosystems; they affect both hydrology and plant growth. Soils are one of the defining features of wetlands, and wetland soil characteristics can take 1Biology and Natural Resources Department, Principia College, Elsah, IL 62028. 2Illinois River Biological Station, Illinois Natural History Survey, Havana, IL 62644. 3Oklahoma Cooperative Fish and Wildlife Research Unit, Oklahoma State University, Stillwater, OK 74078. *Corresponding author - taylorbookout20@gmail.com. Manuscript Editor: Susan Herrick Northeastern Naturalist Vol. 26, No. 2 T. Bookout and G.L. Bruland 2019 393 decades to centuries to develop (Ballantine and Schneider 2009). Wetland soils are differentiated from upland soils by their long periods of saturation during the growing season. The degree of saturation regulates the accumulation of organic matter in the soil, as anaerobic conditions in submerged soils impede the decomposition of organic material (Ballantine and Schneider 2009). Wetland plants are the main primary producers in wetland ecosystems and they have adapted to live in saturated soils. Hydrology plays an important role in defining species composition, with greater saturation favoring obligate wetland species and lower saturation, allowing the establishment of more facultative and upland species. Restored wetlands progress from supporting more facultative annual plant species in early successional stages to more clonal perennials as the restoration progresses (van der Valk 1981). Plant species richness also generally increases, though Matthews et al. (2009) suggested that richness may not be the best metric to assess the success of wetland restoration sites, and that species composition may be more informative. Limiting factors on floristic community development include propagule availability, nutrient input, and hydrology (Matthews and Endress 2010, Matthews et al. 2005). If a site has been farmed for a long period of time (such as the site of this study), has been cut off from historic propagule sources, or has had the topsoil removed, wetland plant species recolonization can be slow if they are not planted as part of the restoration (Seabloom and van der Valk 2003, Steven et al. 2010). Although soil removal can decrease native seed stock, Hausman et al. (2007) found that removal of the upper 10 cm of soil during restoration could improve community quality as facultative propagules were removed and seeds of obligate wetland species buried deeper in the seedbank could reestablish. A common threat to restored wetlands in the Midwest is nutrient input, especially of nitrogen from agricultural runoff, which favors the growth and establishment of non-native plants that prefer disturbed areas with high levels of nutrient availability (Matthews et al. 2009). Plants in both natural and restored wetlands provide important habitat to animals that live in and around wetlands, especially amphibians. Amphibians rely on wetlands as breeding sites and can be good indicators of a wetland’s health (Porej and Hetherington 2005). Urbanization and agricultural development have destroyed and fragmented natural amphibian habitat (Pillsbury and Miller 2008). Wetland restoration efforts have begun to increase amphibian habitat, but little is known about the factors that promote colonization of restored wetlands. Proximity to undisturbed upland habitat and amphibian source populations are positively correlated with rates of amphibian colonization, while fragmentation associated with urban development is negatively correlated with amphibian colonization (Knutson et al. 1999, Lehtinen and Galatowitsch 2001). Hydroperiod, bank slope, and the cover and composition of vegetation are also important factors related to amphibian use of wetlands, especially in early stages of restored wetland development (Porej and Hetherington 2005, Shulse et al. 2012). Different combinations of the factors listed above favor different species by offering longer hydroperiods, warmer areas, or protection, respectively. Northeastern Naturalist 394 T. Bookout and G.L. Bruland 2019 Vol. 26, No. 2 The objective of our study was to assess the ecological status of a restored wetland through the following: (1) comparing the water levels of 3 cells in the restored wetland designed to have different water depths and flood durations; (2) determining trends in soil development among the 3 cells (wetter, intermediate, drier) and across the main hydrologic gradient present within cells; (3) surveying vegetation to determine differences in community composition among the cells; and (4) monitoring relative call frequency to determine amphibian use of the wetland in spring 2016. We hypothesized that, while trends in soil moisture would be observed across the hydrologic gradient, trends in properties such as soil bulk density or organic matter would not be observed across the hydrologic gradient in only the second growing season since restoration. We also hypothesized that plant community composition would be different in the wetter and drier zones within cells, with more obligate species present in the wetter zones. Finally, if wetland hydrology and vegetation were reestablished, we predicted that amphibian species would be present at the site. Methods Site description The study site is located along the lower reach of Piasa Creek in Jersey County, IL. The floodplain adjacent to the creek was farmed continuously for at least 100 y until fall 2013, when the land was acquired by the Great Rivers Land Trust (GRLT). The wetland was restored as mitigation for construction of the new south loading area for America’s Central Port near Granite City, IL, just south of the confluence of the Mississippi River and the Chain of Rocks Canal. Initial surveys in summer 2009 and spring 2010 showed that several depressions with saturated soils existed within the farm field and that the water table was very shallow across the site. Construction of emergent wetland cells and other site improvements, such as the removal of tile drains and invasive plants, began in fall 2013 and concluded in fall 2014. Emergent plants and woody seedlings were planted in spring 2015. A series of 6 emergent wetland cells (~3.34 ha total) were constructed along the depressional area of the floodplain. Three of the cells were designed to be shallow, intermediate buffers between the 3 pool-like cells that were studied. We did not include the 3 intermediate cells as they were dry most of the year except when the entire site was flooded. Cell 1 had the highest elevation with a large shallow basin that held water through most of the growing season. The middle cell (C2) was the largest, with the deepest water bordering a shallow flat that was heavily vegetated. The final cell (C3) was 15 m from the creek and was a shallow depression with a riprap-lined spillway that drained into the stream when water was greater than 30 cm deep. Field sampling We mapped wetter, intermediate, and drier zones in each of 3 of the cells in the restored wetland based on elevation, hydrology, visual survey of vegetation, and standing water, if present (Fig. 1). We used GPS data imported into ArcGIS (Version 10.3; Esri, Redlands, CA) to map these zones. We established 6 random sampling Northeastern Naturalist Vol. 26, No. 2 T. Bookout and G.L. Bruland 2019 395 points in each zone (wetter, intermediate, drier) of each cell (1, 2, 3) for a total of 54 samples. We employed the Sampling Design Tool (Version 1.0, NOAA/NOS/ NCCOS/CCMA/Biogeography Branch, Silver Spring, MD) to generate random points with a minimum 10 m separation (Fig. 1). We exported the coordinates of the sampling points to a GPS for location at the site. We placed a numbered wooden stake at each location to facilitate soil and vegetation sampling. We also placed a staff gauge (height = 1.2 m) in the deepest area of each cell. We recorded maximum water depth in each cell every 2 d in the spring of 2016 (13 March–4 May) and then weekly from 11 May to 17 August. We obtained daily precipitation data from the Weather Underground station at Portage Des Sioux, MO (www.wunderground.com; accessed 1 March 2017), ~6 km from the wetland. Soil We took soil cores ~15 cm from the marker stake, outside the area that was used for the plant survey. We collected the 272-cm3 cores from the upper 15 cm of the soil profile in plastic sleeves and kept the samples in a cooler until they could be processed in the lab. We collected the majority of the cores (42) on the afternoon of 15 April and the last 12 on 16 April (all in the wetter zone so soil moisture was consistent). We placed the cores in pre-weighed aluminum foil boats and weighed them wet and again after they had been dried at 105 °C for at least 24 h. We calculated soil moisture content (SMC) and bulk density (BD) using the wet and dry masses (Bruland and Richardson 2005). We used loss on ignition (LOI) to estimate soil organic matter (SOM) by heating 3–5-g subsamples of the dried soils to 450 °C Figure 1. Site location and site map showing zones delineated in ArcGIS with 54 stratified random sample points. Northeastern Naturalist 396 T. Bookout and G.L. Bruland 2019 Vol. 26, No. 2 for 4 h and weighing the pre- and post-ignited soils with an analytical balance to the nearest 0.0001 g (Bruland and Richardson 2005). Vegetation We sampled 1-m2 quadrats in the wetter, intermediate, and drier zones within each cell (n = 54) at the sampling points described previously. We placed the quadrat frame with the edge centered on the south side of each stake. We conducted the plant survey at the peak of the growing season from 17 to 19 August. We visually assessed percent cover of all species present in each plot using a modified Daubenmire scale (Daubenmire 1959, Matthews and Endress 2010). When multiple vegetation strata were present in the sampling plots, total cover was typically >100%. We categorized plant species by wetland indicator status as obligate wetland (OBL), facultative wetland (FACW), facultative (FAC), facultative upland (FACU), or upland plants (UPL) (USDA NRCS 2018; Table 1). We determined coefficients of conservatism (CoC) for each plant species based on Taft et al. (1997), with lower values indicating species found at weedy, disturbed sites and higher values indicating species intolerant of habitat degradation (Table 1; Matthews et al. 2009). A CoC value of 0 indicates a non-native species (Taft et al. 1997). We calculated the mean CoC for the site based on all species for which CoC values were available. We also calculated the floristic quality index (FQI) with the formula FQI = site mean CoC x S, where S = species richness for the site. Amphibians We conducted frog call surveys (FCS) at dusk within 30 min of sunset in each of the 3 cells every other day starting 13 March and continuing through 10 May. During the summer, we conducted FCS weekly until 17 August. The survey followed the methodology described in Crouch and Paton (2011). Briefly, we observed calls for 3–5 min and quantified relative call frequency (RCF) and intensity using a 4-tier system as follows: (1) no individuals calling (score = 0); (2) calling individuals, but no overlapping (score = 1); (3) some overlap of calls, but still able to discern individual calls (score = 2); and (4) full chorus with full overlap of calls (score = 3). We used Amphibiaweb (https://amphibiaweb.org) for amphibian species identification. We also recorded soil temperature at 5-cm depth, general weather conditions (i.e., cloudy, clear, raining, etc.), time, and surface water depth when we collected the RCF data. Statistical analyses We made visual assessments of data normality in Excel (Version 2016, Microsoft, Seattle, WA) and conducted normality tests in SPSS (IBM SPSS Statistics software, Version 19, IBM, New York, NY). We analyzed soil and vegetation data (i.e., species richness, wetland indicator status, and total cover) with 2-way analysis of variance (ANOVA) in SPSS. We ran a principal components analysis (PCA) using PCOrd (Version 6, MjM Software, Gleneden Beach, OR) to investigate multivariate patterns in the vegetation data and to identify the strongest explanatory variables of those patterns (McCune and Grace 2002). We followed the conventions Northeastern Naturalist Vol. 26, No. 2 T. Bookout and G.L. Bruland 2019 397 of multivariate analyses (McCune and Grace 2002) and removed from the PCA plant species that only occurred in 1 plot. We also conducted an indicator species analysis (ISA) in PCOrd on vegetative data stratified by both hydrologic zone and cell. We assessed associations among soil and vegetative parameters using Pearson and Spearman correlation and associations among RCF and environmental variables with linear regression in SPSS. Results Hydrology Cell 1 (C1) and cell 2 (C2) had surface water at the beginning of data collection on 13 March (Fig. 2). These cells slowly lost water until 31 March, when input from a rain event filled all the cells to their maximum (Fig. 2). Precipitation occured repeatedly during April and May. A rain event of 1.85 cm occurred on 26 May before a dry period with only a few centimeters of precipitation until 30 June. Cells 1 and 2 maintained a relatively stable water level until 4 June when they began to dry. Both C1 and C2 were completely dry by 25 June. The dry period was followed by a large rain event of 5.99 cm on 3 July, that filled the cells, which held water until 19 August when monitoring ended. The larger July rain event was followed by several smaller events of 4.6 cm on 19 July, 1.75 cm on 25 July, 2.51 cm on 12 August, and 1.83 cm on 17 August (Fig. 3). Smaller rain events of 1.5 cm and less occurred throughout July and August. Cell 3 (C3) had a different hydroperiod—there was no surface water for 53% of the study period. This cell’s water level pulsed with large rain events, generally drying in 9–11 d if no other rain events occurred (Fig. 2). Cell 3 reached a maximum Figure 2. Standing water depths of the 3 wetland cells (C1, C2, and C3) in the restored wetland. The water levels of C1 and C2 followed a simler pattern while C3 pulsed with rainfall inputs before drying. Northeastern Naturalist 398 T. Bookout and G.L. Bruland 2019 Vol. 26, No. 2 depth of 0.5 m on 31 March and 25 July, with smaller pulses of 0.35 m on 12 April and 0.29 m on both 2 May and 9 July (Fig. 2). Soil Mean SMC was 20.5% across all sampling locations and varied from 17.2% to 24.2%. Across cells, mean SMC values were significantly higher in C1 (21.4%) than in C2 (19.9%) or C3 (20.3%) (Fig. 4). Cells 2 and 3 were not significantly different from each other (Fig. 4). Mean SMC generally increased from the drier to wetter zones; drier and wetter zones were significantly different, whereas th intermediate zone overlapped with both the wetter and drier zones (Fig. 5). Mean BD was 1.29 g cm-3, varied from 1.07 g cm-3 to 1.49 g cm- 3, and was not significantly different among either cells or zones. Mean SOM was 2.69%, varied from 1.41% to 6.31%, and showed no significant differences across cells or hydrologic zones. Vegetation Of the 34 species planted as part of the initial restoration, we observed 16 (47%) in the vegetative sampling at the end of the second growing season (Table 1). Across the 54 plots sampled, we identified 34 of 46 observed plant species to the species level. The most common species were grasses such as Echinochloa spp. (barnyard grasses; 25 plots), Setaria spp. (foxtail grasses; 19 plots), and Digitaria spp. (crabgrasses; 18 plots). We observed 10 species that only occurred in 1 plot including Nymphaea odorata (Fragrant Water-lily), Penthorum sedoides (Ditch Stonecrop), Potamogeton illinoensis (Illinois Waterweed), Potamogeton pusillus (Baby Pondweed), and Rumex crispus (Curly Dock). We were unable to identify 7 Figure 3. Daily rain totals recorded in Portage Des Sioux at the Weather Underground weather station from 23 May–1 September 2016. Northeastern Naturalist Vol. 26, No. 2 T. Bookout and G.L. Bruland 2019 399 forb species, 2 grass species, and 2 sedge species. Of the identified species, 41.2% (14) were OBL, 26.5% (9) were FACU, 17.6% (6) were FACW, 11.8% (4) were FAC, and 2.9% (1) were UPL (Table 1). The overall mean CoC for the site was 3.1. Twelve individual species had CoCs ≤4, while 7 species had CoCs >4 (CoCs could not be determined for all species; Table 1). Species observed with high CoCs (>5) included Fragrant Water-lily (CoC = 6), the 2 Potamogeton species (CoC = 7), Sagittaria graminea (Grass-leaved Arrowhead; CoC = 7), and Eleocharis palustris (Great Spikerush; CoC = 8) (Table 1). The FQI for the site was 20.7. Figure 4. (A) Mean (± 1 standard error [SE]) soil moisture content (SMC) across the 3 wetland cells. Bars with different letters are significantly different. SMC in C1 was significantly higher than SMC in C2 and C3. (B) Mean (± 1 SE) SMC across the 3 hydrologic zones. Bars with different letters are significantly different. SMC was significantly higher in the wetter zone than the drier zone. The intermediate zone was not significantly dif ferent from either. Northeastern Naturalist 400 T. Bookout and G.L. Bruland 2019 Vol. 26, No. 2 Table 1. Characteristics of vegetative species planted at the study site and observed in the 2nd growing season since restoration. The table lists scientific name, common name, initial planting status (1 if used in the initial planting, 0 if not), presence during the second growing season (1 = yes, 0 = no), wetland indicator status (OBL, FACW, FAC, FACU, UPL; USDA NRCS [2018]), and coefficient of conservatism (CoC; Taft et al. 1997) for each species. Some species (denoted with asterisks [*] in CoC column) observed in this study that were only identified to the genus level do not have a CoC reported in Taft et al. (1997). [Table continued on following page.]. Presence Wetland Initially 2nd indicator Scientific name Common name planted season status CoC Acer spp. Maple 0 1 FAC * Acorus calamus L. Flag Root 1 1 OBL 4 Alisma subcordatum Raf. American Water-plantain 1 0 OBL 2 Ambrosia spp. Ragweed 0 1 FACU * Amsonia tabernaemontana Walter Blue Star 1 0 FACW 6 Asclepias incarnata L. Swamp Milkweed 1 0 OBL 4 Bacopa rotundifolia (Michx.) Wettst. Water Hyssop 1 1 OBL 5 Bidens aristosa (Michx.) Britton Swamp Marigold 1 1 FACW 1 Bidens frondosa L. Common Beggar’s Tick 1 0 FACW 1 Boltonia decurrens Illinois False Aster 1 1 FACW 4 (Torr. & A. Gray) Alph. Wood Carex comosa Boott Bristly Sedge 1 0 OBL 6 Carex cristatella Britton Crested Oval Sedge 1 0 FACW 3 Carex grayi Carey Common Bur Sedge 1 0 FACW 6 Carex praegracilis W. Boott Expressway Sedge 1 1 FACW * Carex vulpinoidea Michx. Brown Fox Sedge 1 0 FACW 3 Cephalanthus occidentalis L. Buttonbush 1 0 OBL 4 Chamaecrista fasciculata (Michx.) Partridge Pea 1 1 FACU 0 Greene Chelone obliqua speciosa Pennell & Pink Turtlehead 1 0 OBL 8 Wherry Conyza canadensis (L.) Cronquist Horseweed 0 1 FACU 0 Cyperus strigosus L. Long-caled Nut Sedge 0 1 FACW 0 Digitaria spp. Crabgrass 0 1 FACU * Echinochloa spp. Barnyard Grass 0 1 FACW 0 Echinodorus berteroi (Spreng.) Fassett Lance-leaved Burhead 1 0 OBL 6 Eleocharis obtusa (Willd.) Schult. Blunt Spikerush 1 1 OBL 2 Eleocharis palustris (L.) Roem. & . Great Spikerush 1 1 OBL 8 Schult Eleocharis parvula (Roem. & Schult.) Dwarf Spikerush 1 1 OBL 0 Link ex Bluff, Nees & Schauer Eryngium yuccifolium Michx. Rattlesnake Master 1 0 FAC 7 Humulus japonicus Siebold & Zucc. Japanese Hops 0 1 FACU 0 Nuphar lutea advena (Aiton) Kartesz Spatterdock 1 0 OBL 6 & Gandhi Nymphaea odorata Aiton Fragrant Water-lily 1 1 OBL 6 Oxalis spp. Wood Sorrel 0 1 FACU * Peltandra virginica (L.) Schott Arrow Arum 1 0 OBL 8 Penthorum sedoides L. Ditch Stonecrop 1 1 OBL 2 Phyla lanceolata (Michx.) Greene Fog Fruit 1 0 OBL 1 Polygonum cespitosum Blume Creeping Smartweed 0 1 FAC 0 Polygonum hydropiperoides Michx. Wild Water-pepper 0 1 OBL 4 Northeastern Naturalist Vol. 26, No. 2 T. Bookout and G.L. Bruland 2019 401 Cell 1 and C2 had a species richness of 24 species, while C3 had a richness of 18 species. The plant community of C1 was composed of 37.5% (9) OBL, 29.2% (7) FACU, 20.8 (5) FACW, and 12.5% (3) FAC. C2 had 50.0% (12) OBL species, 29.2% (7) FACU, 16.7% (4), FACW, and 4.2% (1) FAC. C3 had 33.3% (6) FACU, 33.3% (6) FACW, 22.2% (4) OBL, 5.6% (1) FAC, and 5.6% (1) UPL. In the PCA of the relative cover of 28 recurring species, axis 1 accounted for 31.5%, axis 2 accounted for 17.9% , and axis 3 accounted for 10.2% of the variance, respectively (Fig. 5). Axis 1 covered a gradient from obligate wetland species on the left to upland species on the right of the biplot (Fig. 5). Axis 2 contained mostly facultative wetland species. Plots from the wetter zone loaded mostly on the left side of the biplot with obligate species, drier plots grouped around the upland species, and intermediate plots were scattered between the 2, with a grouping around facultative species in axis 2 (Fig. 5). Indicator species analysis identified various significant indicator species across the wetland (Table 2). For example, Bacopa rotundifolia (Water Hyssop) was a significant indicator of the wetter zone (P = 0.001) and was only found in the wetter zone (Table 2). Cyperus strigosus (Long-scaled Nut Sedge), which occurred mainly in the intermediate zone and was rare in the drier zones, was the strongest indicator of the intermediate zone (P = 0.01; Table 2). The invasive Humulus japonicus (Japanese Hops) was an indicator of the drier zone (P = 0.01); Table 2) and was found only in this zone. Amphibians We observed 1 salamander species: Ambystoma texanum Matthes (Smallmouth Salamander). We also identified 10 species of frogs and toads during the 30 call Table 1, continued. Presence Wetland Initially 2nd indicator Scientific name Common name planted season status CoC Polygonum pensylvanicum L. Pinkweed 0 1 FACW 1 Polygonum spp. Smartweed 0 1 FACW * Pontederia cordata L. Pickerelweed 1 0 OBL 8 Potamogeton illinoensis Morong Illinois Pondweed 1 1 OBL 7 Potamogeton pusillus L. Baby Pondweed 1 1 OBL 7 Rotala ramosior (L.) Koehne Wheelwort 0 1 OBL 4 Rumex crispus L. Curly Dock 0 1 FAC 0 Sagittaria graminea Michx. Grass-Leaved Arrowhead 1 1 OBL 7 Sagittaria latifolia Willd. Common Arrowhead 1 1 OBL 4 Setaria spp. Foxtail 0 1 FACU * Sium suave Water Parsnip 1 0 OBL 5 Solidago spp. Goldenrod 0 1 FACU * Sparganium eurycarpum Engelm. Common Bur-reed 1 1 OBL 5 Thalia dealbata Fraser ex Roscoe Powdery Thalia 1 0 OBL 5 Tridens flavus (L.) Hitchc. Common Purpletop 0 1 UPL 1 Trifolium spp. Clover 0 1 FACU * Typha spp. Cattail 0 1 OBL * Northeastern Naturalist 402 T. Bookout and G.L. Bruland 2019 Vol. 26, No. 2 Figure 5. A principal components analysis biplot of the vegetation data from the restored wetland. Axis 1 accounted for 31.5% variance, Axis 2 = 17.9% of variance, and Axis 3 accounted for 10.2% of variance (total = 59.6%) of variance in vegetation cover data. These 3 axes were significant according to the broken-stick eigen-value test. The dot and dash ellipse on the left shows the location of the wetter plots in ordination space. The dashed line on the right surrounds the drier plots in the plot area. Table 2. Results of the indicator species analysis of the vegetation cover data across hydrologic zones. Mean IV and SD as compared to randomized groups of plants. Species in the table were significant indicators of a zone at a P = 0.05. Observed indicator Species Zone value (IV) Mean IV SD P-value Bacopa rotundifolia Wetter 38.9 10.8 4.53 0.001 Cyperus strigosus Intermediate 22.9 10.8 4.53 0.010 Humulus japonicus Drier 27.8 8.8 4.26 0.010 Rotala ramosior Wetter 26.4 13.0 4.64 0.030 Solidago spp. Wetter 22.2 8.2 3.94 0.030 Typha spp. Wetter 33.3 9.8 4.42 0.003 surveys: Acris crepitans Baird (Northern Cricket Frog), Anaxyrus americanus (Holbrook) (American Toad), Anaxyrus fowleri (Hinkley) (Fowler’s Toad), Hyla versicolor LeConte (Grey Treefrog), Pseudacris crucifer (Wied-Neuwied) (Spring Northeastern Naturalist Vol. 26, No. 2 T. Bookout and G.L. Bruland 2019 403 Peeper), Pseudacris illinoensis Smith (Illinois Chorus Frog), Pseudacris trisariata Wied-Neuwied (Western Chorus Frog), Rana catesbeiana (Shaw) (American Bullfrog), Rana clamitans (Latreille) (Green Frog), and Rana sphenocephala (Cope) (Southern Leopard Frog). The dominant calling species shifted from the Spring Peeper in the early spring to Northern Cricket Frog later in the sampling period. Relative CF data for each cell over time showed that cells with surface water had similar RCF on the same night. We observed a strong linear relationship between soil temperature and RCF for C1 with an R2 of 0.63. Cell 3 had periods of no recorded calls when the cell had no surface water (Figs. 2, 6). The strong relationship of RCF to water depth was fit with a linear regression and had an R2 = 0.79 reinforcing the importance of hydrology on the presence of amphibians. Discussion Hydrology Hydrology is important in wetland development, separating uplands from aquatic ecosystems, and is a key factor in wetland delineation (Mitsch and Gosselink 2015). Hydrology affects soil development, plant community composition, and amphibian usage of wetlands. The successful restoration of wetland hydrology results in appropriate habitat for wetland biota. The diverse cell designs of the studied wetland provided a wide range of habitat and variability within the site. For example, cells 1 and 2 provided more deep-water habitat for water-sensitive plants and amphibians, while also having shallow areas that dried and allowed facultative wetland species to establish. Cell 3 held less surface water, limiting the species composition to more facultative wetland species that did not require surface water for long periods. This cell had a more ephemeral nature than C1 and C2, drained at a greater rate, and spent more than half of the growing season Figure 6. C1 and C2 had similer RCF at the same times while C3 had many period with no calls due to a lack of water. When water was present in C3, RCF spiked before the cell dried. Northeastern Naturalist 404 T. Bookout and G.L. Bruland 2019 Vol. 26, No. 2 without surface water. There are 3 possible explanations for this pattern: (1) C3 has the lowest elevation, and water from this cell readily drains into the adjacent creek; (2) coarser-textured soils may be present closer to the stream and are more prominent in C3; or (3) there may be a relic drain tile in C3 that is facilitating accelerated drainage of this area. Further soil analysis could determine if the soil texture is indeed coarser than other parts of the wetland, thus allowing faster drainage. Also, the spillway out of the wetland could be raised to hold water at a greater depth. Soil Soils are an important component of wetland ecosystems, retaining water, providing a growth medium for plants, and habitat for a variety of other biota. SMC trends supported our hypothesis that we would observe differences in soil moisture across the hydrologic gradient, with significantly higher SMC values in the wetter zones and lower values in the drier zones. As expected, BD and SOM did not exhibit any significantly different values across the zones as soils were still developing in the second growing season after wetland restoration. Soil development can take long periods of time, and many restoration projects take at least 15 y before any significant changes are observed (Ballantine and Schneider 2009). Removal of topsoil and earthmoving may have mixed soils both vertically and horizontally at the site, and such newly formed soil profiles would require more time to develop wetland characteristics (Bantilan-Smith et al. 2009). Vegetation Vegetation is important in wetlands as habitat and is a primary component of the nutrient cycle. Plants provide food for insects, birds, and juvenile amphibians while adding organic matter to the soil. During the spring, we observed large flocks of ducks at the wetland and noted Spizella pusilla (Wilson) (Field Sparrow) among the vegetation at the site throughout the summer. Wetland plants also provide habitat for Agelaius phoeniceus (L.) (Red-wing Blackbird), which we observed nesting in the small Typha spp. (cattails) stand in C1, and frogs and toads used the vegetation during mating activities. Shulse et al. (2012) found that vegetation was an important factor in the colonization of recently constructed wetlands by amphibians, providing cover and a substrate to attach egg masses. Hydrological stability is important in plant community development, affecting both community composition and growth (Matthews et al. 2009). The high number of OBL species we observed (41.2%) suggests that the wetland is developing well within a majority of the cells. This percentage is comparable to planted restored wetlands in Indiana that had ~50% OBL species (Hopple and Craft 2013). In the study site, C1 and C2 showed more stable hydrology and lacked surface water for only a week during the growing season (Fig. 2). Both C1 and C2 supported more OBL wetland species than C3. Cell 2 had the deepest water levels (Fig. 3) and contained the largest number of OBL species (12), while C3 had the fewest OBL species (4) and was dry on 53% of the observation days. Cell 3 also had the highest proportion of FACW and FACU species, with each composing Northeastern Naturalist Vol. 26, No. 2 T. Bookout and G.L. Bruland 2019 405 33.3% (6) of the 18 species identified in C3. The dominance of facultative species is likely the result of the more episodic hydrological cycle present in C3 (Fig. 2) and the demands that periodic flooding place on the plant community. In terms of the vegetative hypothesis, it appeared that there was greater variability among the 3 cells driven by the differential hydrology, rather than within cells across the main hydrologic gradients. The effect of hydrology on community composition was clear in the indicator species analysis. Water Hyssop is an OBL species that was an indicator of the wetter zone and was only found in this zone. It was found primarily (71%) in C2 which had the most extensive wetter zone and only once in C1 and C3. Long-scaled Nutsedge, a FACW plant, was the strongest indicator of the intermediate zone, and was found in the intermediate and drier zones. It occurred mostly in C1 (71%), which had the largest area of intermediate zone of the sampled cells (Fig. 1). The invasive FACU Japanese Hops was an indicator of the drier zone and was found only in this zone (Table 2). The clear differentiation of these species is important in understanding site development and shows that a hydrologic gradient can establish quickly, with a profound effect on the community composition of the wetland. The second vegetative hypothesis (plant community composition will be different across zones, with more obligate species present in the wetter zones) was partially supported by the results of the indicator species analysis. The intensive planting of desired species early in the restoration has helped C1 and C2 develop a community more representative of a later successional wetland, as more clonal perennials like Water Hyssop were established than early annuals in the wetter zones. Cell 3, with its drier hydrology, was dominated by annuals (i.e., Polygonum spp. [smartweeds]) and early colonizers as the planted wetland species struggled to compete with facultative species that can better survive the fluctuations in hydrology. We observed less than half (47%) of the planted species in the second growing season, though some might have been present at the site, but did not appear in our randomly placed quadrats. Alternatively, some of the more sensitive species may not have survived the first 2 years following restoration. Even so, the mean CoC for the study site of 3.1 was comparable to, if not slightly higher than, mean CoCs reported for restored (2.9) and natural wetlands (3.4) in Indiana (Hopple and Craft 2013), as well as mean CoCs for restored (2.2) and natural wetlands (2.3) in Illinois (Matthews et al. 2009). The FQI for the site (20.7) was also comparable to, if not slightly higher than the FQI range for natural and restored wetlands in Indiana (12–21) and mean FQI values for restored (20.7) and natural wetlands in Illinois (12.5) (Matthews et al. 2009). The high plant species richness of this site (46 species) is promising and was on par or higher than species richness values reported in the literature from other restored wetlands. For example, DeBerry and Perry (2004) found 38 species in a restored wetland in Virginia after its second growing season. Likewise, Hopple and Craft (2013) identified a mean of 34 species in restored floodplain wetlands in Indiana with species composition similar to that at our study site. Northeastern Naturalist 406 T. Bookout and G.L. Bruland 2019 Vol. 26, No. 2 Amphibians The presence and abundance of amphibians in a restored wetland is an indicator of the ecological success of that restoration. Amphibians rely on pools and ponds for the development of their young and adjacent upland habitats in which to forage (Todd et al. 2009). The presence of 11 amphibian species in the restored wetland supported our 3rd hypothesis—if wetland hydrology and vegetation were reestablished, we predicted that amphibian species would be present at the site. This result was notable, as an average of only 3.6 amphibian species per site was found in restored wetlands studied in Michigan by Lehtinen and Galatowitsch (2001). Porej and Hetherington (2005) found an average of 4.2 species per site in restored emergent wetlands in Ohio that were an average of 2 y old. A previous study of amphibians in smaller natural wetlands on the nearby Principia College campus showed that frog species richness in these sites was 5–6 species (Klehm 2016), although the restored wetland in our study was larger, located in a floodplain, and less isolated than the natural wetlands in that study. We also observed at the restored wetland many of the same species found in the nearby isolated natural wetlands (Klehm 2016). The proximity to Piasa Creek and the presence of an established pond nearby may have accelerated the development of the diverse amphibian community in such a short period of time. Waterbodies need to have surface water present for several months to ensure that tadpoles have time to metamorphose. Of the 3 cells, C1 and C2 offered the most stable hydroperiod and had the highest total call frequency (Figs. 2 and 6). Both cells held water for 104 consecutive days from the beginning of the study period, long enough to allow many of the tadpoles to metamorphose (Semlitsch et al. 2000, Wilbur 1977). Cell 3 had episodic flooding followed by long periods of drying. This unstable hydroperiod attracted some species for short periods of time, but no eggs or tadpoles survived the dry periods. Frogs and toads responded positively to soil temperature, with higher call frequency associated with warmer temperatures. Amphibians are poikilothermic and their activity is limited by cooler temperatures. The most limiting factor on frog calls was the presence or absence of water seen in C3. When C3 had water, it had similar RCF to the other cells during the same period, but we heard no calls when water was absent (Fig. 6), demonstrating the importance of the hydroperiod on amphibian usage of a wetland. It is important to note that not all amphibians rely on pools and ponds. For example, while all the amphibian species observed in this study use open water areas, there are likely various species of salamanders in the general vicintiy that do not use open water at all. Conclusions The overall development of the restored wetland in its second growing season was promising because of the establishment of a diverse and species-rich (46 species) wetland plant community, extensive use by a surprisingly diverse (11 species) community of amphibians, and signs of wetland soil development across the site. All of the cells had surface water more than 25% of the growing season, which Northeastern Naturalist Vol. 26, No. 2 T. Bookout and G.L. Bruland 2019 407 classifies them as wetlands according to the US Army Corps of Engineers Wetlands Delineation Manual (1987) guidelines. Although hydric soil properties are not strongly visible yet, there are some small-scale trends when viewed spatially. The high richness of the wetland plant community and the ratio of obligate to facultative wetland species is promising for a young wetland restoration site. In broader terms, our results suggest that the extra cost and effort of establishing cells with different depths, and subsequently hydroperiods, in a restored wetland resulted in a more diverse vegetative community than if all cells had the same depths or if no wetland cells had been established. More diverse vegetative communities in restored wetlands can lead to greater diversity of bird, anuran, and other wildlife species. The diverse amphibian community and the presence of the Illinois Chorus Frog should be confirmed with a more extensive amphibian study. If Illinois Chorus Frogs persist, this site could become an important breeding site for this threatened species if it is managed properly. Cell 3 is of some interest because it has a more periodic hydroperiod resulting in fewer OBL wetland plants and lower use by amphibians. While this variability adds to the diversity of the site, the establishment of more upland and facultative upland species in C3 may allow them to outcompete wetland species. The dominant coverage of smartweeds in C3 is also concerning because some of them are non-native, though they are early colonizers and may be replaced if the hydrology changes and more wetland species are established. Given these observations, it appears that C3 is on a different trajectory of successional development than the other cells. The hydrology of C3 may be shifted by slightly raising the outlet so more water is retained or by removing any relict tile drains. This finding also raises larger questions about successional trajectories within restored wetlands and what to do when some areas within a site appear to be trending in a more terrestrial than wetland direction. Further research observing the progression of this wetland and the development of the soils and biotic communities is needed to better assess the future success of the wetland design. Acknowledgments We thank Principia College for financial support for this study and Great Rivers Land Trust for access to the study wetland. We also acknowledge 2 anonymous reviewers and the associate editor for their constructive comments and feedback on the manuscript. Literature Cited Ballantine, K., and R. Schneider. 2009. Fifty-five years of soil development in restored freshwater depressional wetlands. Ecological Applications 19:1467–1480. Bantilan-Smith, M., G.L. Bruland, R.A. MacKenzie, A.R. Henry, and C.R. Ryder. 2009. A comparison of the vegetation and soils of natural, restored, and created coastal lowland wetlands in Hawai‘i. Wetlands 29:1023–1035. DOI:10.1672/08-127.1. Bruland, G.L., C.J. Richardson. 2005. Hydrologic, edaphic, and vegetative responses to microtopographic reestablishment in a restored wetland. Restoration Ecology 13:515–523. Northeastern Naturalist 408 T. Bookout and G.L. Bruland 2019 Vol. 26, No. 2 Crouch, W.B., III, and P.W.C. Paton. 2002. Assessing the use of call surveys to monitor breeding anurans in Rhode Island. Journal of Herpetology. 36(2):185–192. DOI:10.2307/1565990. Daubenmire, R. 1959. A canopy-coverage method of vegetational analysis. Northwest Science 33:43–64. DeBerry, D.A., and J.E. Perry. 2004. Primary succession in a created freshwater wetland. Castanea 69:185–193. Hausman, C.E., L.H. Fraser, M.W. Kershner, and F.A. de Szalay. 2007. Plant community establishment in a restored wetland: Effects of soil removal. Applied Vegetation Science 10:383–390. Hopple, A., and C. Craft. 2013. Managed disturbance enhances biodiversity of restored wetlands in the agricultural Midwest. Ecological Engineering 61:505–510. Klehm, C. 2016. Amphibian breeding habitat use at Principia College. Unpublished Senior Thesis. Principia College, Elsah, IL. Knutson, M.G., J.R. Sauer, D.A. Olsen, M.J. Mossman, L.M. Hemesath, and M.J. Lannoo. 1999. Effects of landscape composition and wetland fragmentation on frog and toad abundance and species richness in Iowa and Wisconsin, USA. Conservation Biology 13:1437–1446. Lehtinen, R.M., and S.M. Galatowitsch. 2001. Colonization of restored wetlands by amphibians in Minnesota. American Midland Naturalist 145:388–396. Matthews, J., and A. Endress. 2008. Performance criteria, compliance success, and vegetation development in compensatory mitigation wetlands. Environmental Management 41:130–141. Matthews, J.W., and A.G. Endress. 2010. Rate of succession in restored wetlands and the role of site context. Applied Vegetation Science 13:346–355. Matthews, J.W., P.A. Tessene, S.M. Wiesbrook, and B.W. Zercher. 2005. Effect of area and isolation on species richness and indices of floristic quality in Illinois, USA wetlands. Wetlands 25:607–615. Matthews, J.W., A.L. Peralta, A. Soni, P. Baldwin, A.D. Kent, and A.G. Endress. 2009. Local and landscape correlates of non-native species invasion in restored wetlands. Ecography 32:1031–1039. McCune, B., and J.B. Grace. 2002. Analysis of ecological communities. Journal of Experimental Marine Biology and Ecology 289:303–305. Mitsch, W.J., and J.G. Gosselink. 2015. Wetlands, 5th Edition. Wiley, Hoboken, NJ. 456 pp. Pillsbury, F.C., and J.R. Miller. 2008. Habitat and landscape characteristics underlying anuran community structure along an urban–rural gradient. Ecological Applications 18:1107–1118. Porej, D., and T.E. Hetherington. 2005. Designing wetlands for amphibians: The importance of predatory fish and shallow littoral zones in structuring of amphibian communities. Wetlands and Ecological Management 13(4):445–455. Seabloom, E.W., and A.G. van der Valk. 2003. Plant diversity, composition, and invasion of restored and natural prairie pothole wetlands: Implications for restoration. Wetlands 23:1–12. Semlitsch, R.D., C.M. Bridges, and A.M. Welch. 2000. Genetic variation and a fitness tradeoff in the tolerance of Gray Treefrog (Hyla versicolor) tadpoles to the insecticide carbaryl. Oecologia 125:179–185. Shirose, L.J., C.A. Bishop, D.M. Green, C.J. MacDonald, R.J. Brooks, and N.J. Helferty. 1997. Validation tests of an amphibian call count survey technique in Ontario, Canada. Herpetologica 53:312–320. Northeastern Naturalist Vol. 26, No. 2 T. Bookout and G.L. Bruland 2019 409 Shulse, C.D., R.D. Semlitsch, K.M. Trauth, and J.E. Gardner. 2012. Testing wetland features to increase amphibian reproductive success and species richness for mitigation and restoration. Ecological Applications 22:1675–1688. Steven, D.D., R.R. Sharitz, and C.D. Barton. 2010. Ecological outcomes and evaluation of success in passively restored southeastern depressional wetlands. Wetlands 30:1129–1140. Taft, J.B., G.S. Wilhelm, D.M. Ladd, and L.A. Masters. 1997. Floristic quality assessment for vegetation in Illinois, a method for assessing vegetation integrity. Erigenia 15:3–95. Todd, B.D., T.M. Luhring, B.B. Rothermel, and J.W. Gibbons. 2009. Effects of forest removal on amphibian migrations: Implications for habitat and landscape connectivity. Journal of Applied Ecology 46:554–561. US Army Corps of Engineers Wetlands Delineation Manual. 1987. Environmental Laboratory Report Number Y-87-1. US Army Engineer Waterways Experiment Station, Vicksburg, MS. USDA Natural Resources Conservation Service (USDA NRCS). 2018. Plants Database. Available online at https://plants.sc.egov.usda.gov/java/. Accessed 17 May 2018. van der Valk, A.G. 1981. Succession in wetlands: A Gleasonian approach. Ecology 62:688–696. Wilbur, H.M. 1977. Density-dependent aspects of growth and metamorphosis in Bufo americanus. Ecology 58:196–200.